Control device, control system, and machine learning device

Abstract

A control device according to the present invention is provided with a data acquisition unit configured to acquire data on at least an operating state of an industrial machine, a learning model storage unit configured to store a learning model in which the value of a setting action for a base speed of a servomotor for peak cut is associated with the operating state of the industrial machine, and a decision making unit configured to settle the setting action for the base speed of the servomotor for peak cut based on the data on the operating state of the industrial machine acquired by the data acquisition unit, by using the learning model stored in the learning model storage unit.

The invention claimed is: 
     
       1. A control device configured to control a peak-cut operation of a motor for peak cut connected to one and the same power supply path with at least one industrial machine, the control device comprising:
 a data acquisition unit configured to acquire data on at least an operating state of the industrial machine; 
 a learning model storage unit configured to store a learning model in which the value of a setting action for a base speed of the motor for peak cut is associated with the operating state of the industrial machine; and 
 a decision making unit configured to settle the setting action for the base speed of the motor for peak cut based on the data on the operating state of the industrial machine acquired by the data acquisition unit, by using the learning model stored in the learning model storage unit, wherein regenerative power is generated by decelerating the motor for peak cut. 
 
     
     
       2. The control device according to  claim 1 , wherein the data acquisition unit further acquires data on the base speed of the motor for peak cut, and the control device comprises a learning unit configured to generate a learning model in which the setting action for the base speed of the motor for peak cut is associated with the operating state of the industrial machine. 
     
     
       3. The control device according to  claim 2 , wherein the learning unit identifies a positive reward if supply of regenerative power from the motor for peak cut is sufficient for electric power consumed in the industrial machine or if the set base speed of the motor for peak cut is low and identifies a negative reward if the supply of the regenerative power from the motor for peak cut is insufficient for the electric power consumed in the industrial machine or if the set base speed of the motor for peak cut is high, and the learning unit generates the learning model based on the value of the reward concerned. 
     
     
       4. The control device according to  claim 1 , wherein the learning model is an action value table stored with the value of the setting action for the base speed of the motor for peak cut in association with the operating state of the industrial machine. 
     
     
       5. The control device according to  claim 1 , wherein the learning model is a neural network formed of a multi-layer structure. 
     
     
       6. A control system in which a plurality of the control devices according to  claim 2  are connected to one another, the control system being configured so that the result of learning by the learning unit is sharable by the control devices. 
     
     
       7. A machine learning device having learned a setting action for a base speed of a motor for peak cut, connected to one and the same power supply path with at least one industrial machine, in control of a peak-cut operation of the motor for peak cut, the machine learning device comprising:
 a learning model storage unit configured to store a learning model in which the value of a setting action for a base speed of the motor for peak cut is associated with an operating state of the industrial machine; and 
 a decision making unit configured to settle the setting action for the base speed of the motor for peak cut based on the data on the operating state of the industrial machine by using the learning model stored in the learning model storage unit, 
 wherein regenerative power is generated by decelerating the motor for peak cut. 
 
     
     
       8. The machine learning device according to  claim 7 , further comprising a learning unit configured to generate a learning model in which the setting action for the base speed of the motor for peak cut is associated with the operating state of the industrial machine.